{"id":"https://openalex.org/W4387687112","doi":"https://doi.org/10.48550/arxiv.2310.08780","title":"\"Im not Racist but...\": Discovering Bias in the Internal Knowledge of Large Language Models","display_name":"\"Im not Racist but...\": Discovering Bias in the Internal Knowledge of Large Language Models","publication_year":2023,"publication_date":"2023-10-13","ids":{"openalex":"https://openalex.org/W4387687112","doi":"https://doi.org/10.48550/arxiv.2310.08780"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.08780","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.08780","pdf_url":"https://arxiv.org/pdf/2310.08780","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2310.08780","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5046056262","display_name":"Abel Salinas","orcid":"https://orcid.org/0009-0008-2871-0051","observed_orcids":["https://orcid.org/0009-0008-2871-0051"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Salinas, Abel","raw_affiliation_strings":["USC Information Sciences Institute"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"USC Information Sciences Institute","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025456644","display_name":"Louis Penafiel","orcid":"https://orcid.org/0000-0003-2494-2978","observed_orcids":["https://orcid.org/0000-0003-2494-2978"]},"institutions":[{"id":"https://openalex.org/I4210118379","display_name":"Aptima (United States)","ror":"https://ror.org/02js2n445","country_code":"US","type":"company","lineage":["https://openalex.org/I4210118379"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Penafiel, Louis","raw_affiliation_strings":["Aptima , Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aptima , Inc","institution_ids":["https://openalex.org/I4210118379"]}]},{"author_position":"middle","author":{"id":null,"display_name":"McCormack, Robert","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210118379","display_name":"Aptima (United States)","ror":"https://ror.org/02js2n445","country_code":"US","type":"company","lineage":["https://openalex.org/I4210118379"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"McCormack, Robert","raw_affiliation_strings":["Aptima , Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aptima , Inc","institution_ids":["https://openalex.org/I4210118379"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002709735","display_name":"Fred Morstatter","orcid":"https://orcid.org/0000-0002-0247-4328","observed_orcids":["https://orcid.org/0000-0002-0247-4328"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Morstatter, Fred","raw_affiliation_strings":["USC Information Sciences Institute"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"USC Information Sciences Institute","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.9667999744415283,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.9667999744415283,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.010200000368058681,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.005900000222027302,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9909999966621399},{"id":"https://openalex.org/keywords/social-bias","display_name":"social bias","score":0.5820000171661377},{"id":"https://openalex.org/keywords/stereotypes","display_name":"stereotypes","score":0.9139999747276306},{"id":"https://openalex.org/keywords/internal-knowledge","display_name":"internal knowledge","score":0.6800000071525574},{"id":"https://openalex.org/keywords/knowledge-representation","display_name":"knowledge representation","score":0.7379999756813049},{"id":"https://openalex.org/keywords/fairness","display_name":"fairness","score":0.4690000116825104},{"id":"https://openalex.org/keywords/transparency","display_name":"transparency","score":0.6420000195503235}],"concepts":[{"id":"https://openalex.org/C2780233690","wikidata":"https://www.wikidata.org/wiki/Q535347","display_name":"Transparency (behavior)","level":2,"score":0.6541587114334106},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.610406756401062},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5988032817840576},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5572352409362793},{"id":"https://openalex.org/C2776035688","wikidata":"https://www.wikidata.org/wiki/Q1606558","display_name":"Affect (linguistics)","level":2,"score":0.5338111519813538},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5115538239479065},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.4542028605937958},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.38227522373199463},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3639669418334961},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.3278025686740875},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2989301085472107},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.18014904856681824},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.11012694239616394},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.10298323631286621},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08478587865829468},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C59822182","wikidata":"https://www.wikidata.org/wiki/Q441","display_name":"Botany","level":1,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.08780","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.08780","pdf_url":"https://arxiv.org/pdf/2310.08780","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2310.08780","doi":"https://doi.org/10.48550/arxiv.2310.08780","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.08780","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2310.08780","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.08780","pdf_url":"https://arxiv.org/pdf/2310.08780","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.7307999730110168}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4387687112.pdf","grobid_xml":"https://content.openalex.org/works/W4387687112.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4416940388","https://openalex.org/W4391632026","https://openalex.org/W7155415802","https://openalex.org/W6989077593","https://openalex.org/W4396889891","https://openalex.org/W4396744498","https://openalex.org/W4407762318","https://openalex.org/W4385507563","https://openalex.org/W4409361130","https://openalex.org/W4386081379"],"abstract_inverted_index":{"Large":[0],"language":[1,19,111],"models":[2,24],"(LLMs)":[3],"have":[4,25],"garnered":[5],"significant":[6],"attention":[7],"for":[8,97],"their":[9,39,42,98],"remarkable":[10],"performance":[11,40],"in":[12,41,90,109],"a":[13,51,68,94],"continuously":[14],"expanding":[15],"set":[16],"of":[17,71,77],"natural":[18,110],"processing":[20,112],"tasks.":[21],"However,":[22],"these":[23],"been":[26],"shown":[27],"to":[28,56,103],"harbor":[29],"inherent":[30],"societal":[31],"biases,":[32],"or":[33],"stereotypes,":[34,73],"which":[35],"can":[36],"adversely":[37],"affect":[38],"many":[43],"downstream":[44],"applications.":[45],"In":[46],"this":[47],"paper,":[48],"we":[49],"introduce":[50],"novel,":[52],"purely":[53],"prompt-based":[54],"approach":[55,65],"uncover":[57],"hidden":[58],"stereotypes":[59],"within":[60,80],"any":[61],"arbitrary":[62],"LLM.":[63],"Our":[64],"dynamically":[66],"generates":[67],"knowledge":[69],"representation":[70],"internal":[72,83],"enabling":[74],"the":[75,81,87],"identification":[76],"biases":[78,88],"encoded":[79],"LLM's":[82],"knowledge.":[84],"By":[85],"illuminating":[86],"present":[89],"LLMs":[91],"and":[92,106],"offering":[93],"systematic":[95],"methodology":[96],"analysis,":[99],"our":[100],"work":[101],"contributes":[102],"advancing":[104],"transparency":[105],"promoting":[107],"fairness":[108],"systems.":[113]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2023-10-17T00:00:00"}